The kidneys remove numerous organic solutes from the plasma. Remarkably little is known, however, about the relation between these solutes and health outcomes in patients with kidney disease. Care is now guided almost exclusively by measurement of only two solutes, creatinine and urea, which are known to be imperfect predictors of clinical events. The proposed study will employ new mass spectrometric methods to identify additional solutes associated with important clinical outcomes. These metabolomic methods allow for the simultaneous measurement of numerous solutes in individual samples. The current proposal will apply these methods to two major clinical problems. The first aim is to identify solutes for which impaired tubular secretion predicts the progression of chronic kidney disease. In focusing on glomerular function, nephrologists have neglected the possibility that impaired tubular secretion predicts the progression of chronic kidney disease (CKD). The candidate has recently employed high resolution untargeted mass spectrometry to identify solutes which are very efficiently secreted by the normal kidneys. The proposed study will use similar methods to interrogate appropriately stored samples from prospective studies of patients with CKD. Untargeted mass spectrometry will identify candidate solutes for which impaired secretion is potentially associated with the rat of progression of CKD. Follow-up studies will establish the chemical identity of solutes of interest. More quantitative assays using tandem mass spectrometry with isotopic dilution will then be employed to confirm the relation of specific solutes to progression in a larger population. Results will test the hypothesis that impaired tubular secretion predicts the risk of kidney disease progression independent of established markers. The second aim is to identify solutes which may cause uremic symptoms. We know remarkably little about which retained solutes cause illness when the kidneys fail. The proposed studies will use untargeted mass spectrometry to interrogate plasma samples from patients maintained on hemodialysis. Untargeted mass spectrometry will identify candidate solutes potentially associated with prominent uremic symptoms. Follow-up studies will establish the chemical identity of solutes of interest. More quantitative assays using tandem mass spectrometry with isotopic dilution will then be employed to confirm the relation of specific solutes to symptoms in a larger population. Results will test the hypothesis that state of the art mass spectrometry can identify retained solutes associated with uremic symptoms. Both aims are designed to yield new chemical knowledge which could facilitate the design of future clinical trials and potentially lead to new methods of treatment.